← Back to Documentation
Models & Providers
Custom Providers
Add OpenAI-compatible providers, private gateways, and model routing rules.
Key Steps
- Configure the provider base URL, credential strategy, and health checks.
- Create aliases that hide backend-specific model IDs from users.
- Use routing and fallback rules to keep approved model access consistent.
Article Scope
This guide focuses on the operational steps needed to run Orchestris with team access, provider control, and predictable usage.
Visual Reference
Provider definition
- Set the base URL and authentication header expected by the provider.
- Name the provider for the business or infrastructure owner, not only the backend technology.
- Run a health check before exposing models to users.
Model aliases
- Create stable aliases for model IDs.
- Use aliases to preserve workflows when backend model IDs change.
- Document which alias is intended for drafting, analysis, coding, or private-data workflows.
Fallback and review
- Plan fallback behavior for provider outages.
- Review cost and latency after routing changes.
- Remove aliases that point to retired models.
Related Topics
Models & Providers
Using AI Models
Help users choose the right models without exposing provider complexity.
Models & Providers
OpenAI Integration
Connect OpenAI models and expose approved options through the provider catalog.
Models & Providers
Anthropic Integration
Connect Anthropic models for teams that need strong long-context and reasoning workflows.